First Stage Analysis of the Energy response and resolution of the Scintillator ECAL in the Beam Test at FNAL, 2008

Similar documents
First Stage Analysis of the Energy Response and Resolution of the Scintillator ECAL in the Beam Test at FNAL, 2008

Construction of the CALICE High Granular Scintillator Based Electromagnetic Calorimeter Prototype and Its Response to Electrons

Status of Simulation Tools for ScECAL of ILD

CALICE Si-W EM Calorimeter: Preliminary Results of the Testbeams 2006

Electromagnetic calorimetry for the ILC

CALICE Test Beam Data and Hadronic Shower Models

CALICE scintillator HCAL

Status of the CALICE analog calorimeter technological prototypes

Validation of Geant4 Physics Models Using Collision Data from the LHC

PoS(TIPP2014)033. Upgrade of MEG Liquid Xenon Calorimeter. Ryu SAWADA. ICEPP, the University of Tokyo

Combined Calorimeters Properties Part 1.

Test setup, APDs,, preamps Calibration procedure Gain monitoring with LED Beam test results Future R&D options

Digital Imaging Calorimetry for Precision Electromagnetic and Hadronic Interaction Measurements

Digital Calorimetry for Future Linear Colliders. Tony Price University of Birmingham University of Birmingham PPE Seminar 13 th November 2013

PERFORMANCE OF THE ATLAS LIQUID ARGON FORWARD CALORIMETER IN BEAM TESTS

Upgrade of the CMS Forward Calorimetry

Dual readout with tiles for calorimetry.

Functions Represent SiPM Response Especially Linear Behavior After Saturation

David Gascón. Daniel Peralta. Universitat de Barcelona, ECM department. E Diagonal 647 (Barcelona) IFAE, Universitat Aut onoma de Barcelona

Hadronic energy reconstruction in the combined electromagnetic and hadronic calorimeter system of the CALICE Collaboration

STATUS OF ATLAS TILE CALORIMETER AND STUDY OF MUON INTERACTIONS. 1 Brief Description of the ATLAS Tile Calorimeter

arxiv: v1 [hep-ex] 18 Feb 2009

arxiv: v1 [physics.ins-det] 23 Oct 2017

G. Gaudio, M. Livan The Art of Calorimetry Lecture V. The state of art Towards ILC calorimetry

A study of silicon sensor for ILD ECAL

PoS(TIPP2014)093. Performance study of the TOP counter with the 2 GeV/c positron beam at LEPS. K. Matsuoka, For the Belle II PID group

Calorimetry in particle physics experiments

Thin Calorimetry for Cosmic-Ray Studies Outside the Earth s Atmosphere. 1 Introduction

Simulation study of scintillatorbased

Characterisation of Silicon Photomultipliers for the T2K Experiment

SciBar and future K2K physics. F.Sánchez Universitat Aútonoma de Barcelona Institut de Física d'altes Energies

1 Introduction. KOPIO charged-particle vetos. K - RARE Meeting (Frascati) May Purpose of CPV: veto Kl

Calorimeter test-beam results with APDs

The Electromagnetic Calorimeter of the HERA-B Experiment

ATLAS Tile Calorimeter performance for single particles in beam tests

NA62: Ultra-Rare Kaon Decays

Simulation and validation of the ATLAS Tile Calorimeter response

Calorimetry I Electromagnetic Calorimeters

Les Premières Données dans ATLAS et le Calorimètre à Argon Liquide

Study of TileCal Sampling Fraction for Improvement of Monte-Carlo Data Reconstruction

Studies of the ATLAS hadronic Calorimeter response to muons at Test Beams

Temperature Dependence Calibration and Correction of the DAMPE BGO Electromagnetic Calorimeter

Commissioning of the ATLAS LAr Calorimeter

Muon reconstruction performance in ATLAS at Run-2

Information about the T9 beam line and experimental facilities

ATLAS Tile Calorimeter Calibration and Monitoring Systems

ABSTRACT RESULTS OF BEAM TESTS OF A PROTOTYPE CALORIMETER FOR A LINEAR COLLIDER

Response curve measurement of the SiPM

ATLAS Hadronic Calorimeters 101

First Data Quality of the CALICE Test Beam 2008

A gas-filled calorimeter for high intensity beam environments

What detectors measure

LHCb Calorimetry Impact

Equalisation of the PMT response to charge particles for the Lucid detector of the ATLAS experiment

High-energy Gamma Rays detection with the AMS-02 electromagnetic calorimeter. F. Pilo for the AMS-02 ECAL Group INFN Sezione di Pisa, Italy

5 Hadronic calorimetry

7th International Conference on New Developments In Photodetection

Digital Hadron Calorimetry for the Linear Collider using GEM Technology

Performance of the MCP-PMT for the Belle II TOP counter

Particle Flow Algorithms

The Compact Muon Solenoid Experiment. Conference Report. Mailing address: CMS CERN, CH-1211 GENEVA 23, Switzerland

The ATLAS Liquid Argon Calorimeter: Construction, Integration, Commissioning Ph. Schwemling on behalf of the ATLAS LAr Group

Calorimeter for detection of the high-energy photons

Electroweak Physics at the Tevatron

Last Lecture 1) Silicon tracking detectors 2) Reconstructing track momenta

arxiv: v2 [nucl-ex] 6 Oct 2008

The reaction p(e,e'p)π 0 to calibrate the Forward and the Large Angle Electromagnetic Shower Calorimeters

PoS(KAON09)023. Beam Hole Photon Veto For J-PARC K O TO experiment. Yosuke Maeda Kyoto University

arxiv: v1 [physics.ins-det] 22 Dec 2013

CsI Calorimeter for KOTO experiment

ATLAS EXPERIMENT : HOW THE DATA FLOWS. (Trigger, Computing, and Data Analysis)

arxiv: v1 [physics.ins-det] 3 Dec 2018 Fast Interaction Trigger for the upgrade of the ALICE experiment at CERN: design and performance

FoCal Project in ALICE. Yota Kawamura for the ALICE FoCal collaboration TCHoU workshop 2018/3/15

9/27 JUNE 2003 SUMMER STAGE PARTICLES REVELATION THROUGH CERENKOV AND SCINTILLATION COUNTER AND THE CEBAF EXPERIMENT

Particle detection 1

The First Large Application of MPPC: T2K Neutrino Beam Monitor INGRID

Adam Para, Fermilab CALOR2010, IHEP, Beijing May 14, 2010 TOTAL ABSORPTION HOMOGENEOUS CALORIMETER WITH DUAL READOUT

Commissioning and Calibration of the Zero Degree Calorimeters for the ALICE experiment

The achievements of the CERN proton antiproton collider

Technological Prototypes and Result Highlights of Highly Granular Calorimeters

HL-LHC Physics with CMS Paolo Giacomelli (INFN Bologna) Plenary ECFA meeting Friday, November 23rd, 2012

ATLAS NOTE October 12, 2009

Photons: Interactions

The Fast Interaction Trigger Upgrade for ALICE

Identifying Particle Trajectories in CMS using the Long Barrel Geometry

Stathes D. Paganis Nevis Laboratories, Columbia University, Irvington NY, 10533, USA (On behalf of the ZEUS Collaboration)

Calibration and Performance of the ATLAS Tile Calorimeter During the LHC Run 2

MINOS. Physics Program and Construction Status. Karol Lang The University of Texas at Austin. YITP: Neutrinos and Implications for Physics Beyond

New Results from the DREAM project

Missing Transverse Energy. Performance with CMS

MEG II 実験の準備状況と 今後の展望 東京大学 素粒子物理国際研究センター 岩本敏幸 他MEG II コラボレーション 2018年9月14日 日本物理学会2018年秋季大会 信州大学 JSPS Core-to-Core Program

Reconstruction in Collider Experiments (Part IX)

THE AMS RICH COUNTER

arxiv: v1 [hep-ex] 6 Jul 2007

V0 cross-section measurement at LHCb. RIVET analysis module for Z boson decay to di-electron

Examples for experiments that can be done at the T9 beam line

CALICE Calorimetry for the International Linear Collider

ATLAS New Small Wheel Phase I Upgrade: Detector and Electronics Performance Analysis

GMp Experiment (E ): An update

ALICE Commissioning: Getting ready for Physics

Transcription:

1 3 4 First Stage Analysis of the Energy response and resolution of the Scintillator ECAL in the Beam Test at FNAL, 8 The CALICE collaboration October 3, 9 5 6 7 8 9 11 1 13 14 This note contains preliminary results of the ScECAL beam test at Fermilab meson test beam line in September 8 and is for the use of members of the CALICE collaboration. Abstract A beam test of the ScECAL prototype module has been performed in the period Aug 19 - Oct 4 8 at FNAL meson beam test facility. This beam test is the first comprehensive beam test for the ScECAL combined with Analog ScHCAL using 1-3 GeV electron, pion and muon beams. As a preliminary result of the first stage analysis, we obtain the ScECAL energy resolution a σ E /E = (1.44 ±.)% (15.15 ±.3)/ E beam (GeV )% for itsthe electron beam, with the quoted uncertainty purely statistic. The deviation from the perfect linear response is found to be less than 6%. 15 16 17 18 19 1 3 4 5 6 7 8 9 3 31 3 33 34 35 36 37 1 Introduction The International Linear Collider (ILC) experiment requires the high precision measurement using the clear initial state and well reconstructed final state of the electron-positron collider. To determine the final state in quarks leptons and gauge bosons, the reconstruction of jets is one of the key issues. The mandatory way to precisely reconstruct jets is combination of calorimetry and tracking, constructing individual particles within jets. This method, called particle flow approach (PFA) requires the fine granular electromagnetic calorimeter. In the Letter of Intent of ILD[1] recently released, the granularity of the electromagnetic calorimeter (ECAL) is required to be finer than 1. 1. segmentation. The Scintillator ECAL (ScECAL) is one of proposed concepts of the fine granular electromagnetic calorimeter for the ILC, which is designed to have effective 1 1 lateral segmentation using 1 4.5 strip scintillator. In order to achieve the required 1 1 lateral segmentation, the strips in odd layers are orthogonal with respect to those in the even layers. The proposed detector is a sampling calorimeter made of mm thick plastic scintillator as the sensitive layers and of 3 mm thick tungsten as the absorber. The scintillation photons are collected by a wavelength shifting (WLS) fiber, inserted along the longitudinal direction of center of each strip scintillator and are readout with a Pixelated Photon Detector (PPD). As the first prototype of the ScECAL, the ScECAL group made a test module which has 9 9 area for one layer and 6 scintillator layers in between 3.5 mm thick tungsten-cobalt absorber layers[, 3, 4, 5]. The total number of radiation lengths is 18.5. This first prototype of the ScECAL was tested using 1-6 GeV/c positron beam at DESY in March 7. The deviation from the linear behavior of energy response is less than 1% indicating the good linearity in this energy region. The stochastic term of the energy resolution curve, 14% shows good energy resolution. However, the 1

38 39 4 41 4 43 44 45 46 47 48 49 5 51 5 53 54 55 56 57 58 59 6 61 6 63 64 65 66 67 68 69 7 71 7 73 74 75 76 77 78 79 8 81 8 constant term of the energy resolution curve is 3%. One of the reason of this rather large constant term is from the shower leakage. The other reason is the non-uniformity of response in each strip. In the first beam test we found that the constant term depended on the beam position when the strips had non-uniformity[3]. At the center of the ScECAL surface, the near PPD sides and the other sides of scintillators are adjacently aligned. Therefore, the sensitivity is drastically changed in a small area, when we use the strip scintillators whose sensitivity decreases as the position is away from the PPD. On the other hand, on a position which defined as.5 away from the center in both horizontal and vertical direction, the sensitivity is stable with positions even if we use such scintillators. The former is called central region and the latter is called uniform region. As the beam aimed in the center region, the effect of non-uniformity of scintillator response to the energy resolution is larger, while the effect of non-uniformity is minimized on the uniform region. According to these results of the first beam test, we built the second ScECAL prototype with many improvements and modifications. Although the scintillator uniformity improved for the strips used in the new prototype, only its effect on the results are presented and discussed in this note. We will discuss the method we developed to increase the uniformity in a future note. The second prototype has been doubled in lateral size with respect to the prototype tested at DESY, having a lateral size of 18 18. The number of layers has been increased to 3 layers so that the number of radiation length of this calorimeter is 1.3 (Hereafter we denote the ScECAL prototype as the second prototype.). We then, had two more comprehensive beam tests for this second ScECAL prototype, in Sep. 8 and May 9 at FNAL. The ScECAL prototype was exposed to electron and hadronic beams up to 3 GeV/c together with the analog scintillator hadron calorimeter (AHCAL) and the Tail Catcher in order to evaluate the combined performance of calorimeters. This note briefly explains the analysis to evaluate the linearity and resolution of the energy measurement by the nd ScECAL prototype using only electron beams data taken at FNAL in September 8. The results in this note are very preliminary, because they are estimated without any systematic uncertainty. More detail and comprehensive analysis results will follow in consecutive analyses near future. The outline of this note is as follows. In Section the detailed description of the ScECAL tested at FNAL is given. The beam line and detector setup are described in Section 3, the analysis including the calibration of each strip, PPD saturation correction and event selection is described in Section 4. In Section 5 the results on the linearity and the energy resolution are described and are the followed by the discussion about these results in in Section 6. Finally, summary and outlook are presented in Section 7. The ScECAL Second prototype The structure of the ScECAL test module is shown in Figure 1. The module consists of 3 pairs of scintillator and absorber layers with thickness of 3. mm and 3.5 mm respectively. The absorber layer consists of 88% tungsten, 1% cobalt and.5% carbon, and has Moliere radius of mm. Each scintillator layer has 18 4 scintillator strips whose size is 4.5 1.. In successive scintillator layers, the strips are alternately aligned vertically ( X layers) and horizontally ( Y layers). The coordination in this note is the right hand system and z direction is paralleled with beam direction. Regarding to the PPDs, the 16-picel MPPCs by Hamamatsu KK were chosen. The gain, capacitance, dark noise and breakdown voltage were measured for them. They were then soldered on a flat cable, and mounted in a housing in the end of each scintillator strip. Signals from the MPPCs are fed into the readout baseboard through the flat cable. Figure

Figure 1: Structure of the ScECAL test module in Sep 8 at FNAL. a shows a housing for a MPPC, b shows a scintillator strip hermetically covered by KIMOTO reflector, and c shows a hole to introduce LED light for the gain monitoring system. Table 1: A list of trigger systems for particles including the pressure of C erenkov counter. Particle p(gev/c) muon electron electron electron electron electron electron electron 3 1 3 6 1 16 5 3 Trigger C erenkov pressure 345 hpaa (outer) 345 hpaa (outer) 345 hpaa (outer) 138 hpaa (outer) 138 hpaa (outer) 138 or 3 hpaa (outer) 3 hpaa (outer) 88 shows a photograph of the ScECAL test module mounted on the beam line and in front of the AHCAL to test the combining performance of total calorimeter including the Tail Catcher. The size of the entire module is 18 18 6 and the total number of readout channels is 16. The WLS fibers to collect the scintillation photons are Kuraray Y-11 double clad type. The strip scintillators were made by extrusion method in Kyungpook National University (KNU). Each strip scintillator is hermetically covered with the reflecting film made by KIMOTO co. 89 3 83 84 85 86 87 9 91 9 93 94 95 96 Beams and the experimental setup The beam test has been performed at the MT6 experiment area in the Meson Test Beam Facility (MTBF) of FNAL. Electron and charged pion beams with an energy between 1 GeV to 3 GeV were used. The muon beams at 3 GeV were also provided for MIP calibration. The setup of the beam line is shown in Figure 3. We used some combinations of trigger counters including the C erenkov counter placed in upstream of the beam site. A pair of trigger for muon run and trigger for pion and electron beam runs were used. The trigger combinations for the particles are listed in Table 1. 3

Figure : Photograph of the ScECAL test module in front of AHCAL. First absorber layer of the ScECAL can be seen on the photo. 97 98 99 1 3 4 5 6 7 8 9 1 111 11 113 114 115 116 117 118 119 The electron beam momentum was tuned at 1, 3, 6, 1, 16, 5, 3 GeV/c. Momentum spread of the provided beams is.%[6]. Beams were injected into the central region and uniform region, according to the mention in Section 1,to confirm the improvement in uniformity of scintillator response. 4 Analysis 4.1 Detector calibration with MIP-like particles The first step in the analysis is the calibration of the strips using muons in order to measure the energy deposits by a Minimum Ionizing Particles (MIPs). This analysis gives the converting factor between ADC counts and the number of MIPs. The muon-tuned beam contains almost no electrons or pions because of the iron dumper put on beam line upstream of our experiment site. Therefore, the MIP events are only required to have the same X and Y hit position of channels at least in X and Y layers, respectively. Decision of the hit is defined as that the size of signal from the mean of pedestal is larger than three times of standard deviation of pedestal. On each channel the MIP response is obtained as corresponding charge in ADC counts between the pedestal peak and the most probable value (MPV) of the distribution of MIP events. The MPV of MIP signal is measured fitting the distribution with a Landou function convoluted with a Gaussian function. Four typical ADC distribution of MIP events are shown in Figure 4. The distribution of MIP constants of 16 channels is shown in Figure 5. The averaged MIP constant of 16 channels is 16 ADC counts. It is significantly larger than the corresponding standard deviation of the pedestal which is 15 ADC counts. Statistic uncertainty of the MIP response on each channel is less than.6% even in the worst channels. There are five dead channels due to broken photo-sensors or too high levels of background noise. The dead channels are either in the half of the ScECAL which is furthest downstream or on the side of the calorimeter. Therefore, the contribution of the dead channels to the energy measurement is negligible. 4

Figure 3: Configuration of the beam line at the MT6 in FNAL. a shows the ScECAL, b shows AHCAL and c shows the Tail catcher. 11 1 13 14 15 16 17 4. The correction of MPPC saturation The MPPC response has a saturation behavior according to its intrinsic property. Thus the saturation correction for each MPPC must be implemented. We have measured the saturation effect of the MPPC using a simple bench test in advance to perform the beam test. Figure 6 shows the set up of the experiment. One of our strips which was the same as the one we used in the prototype module had been illuminated with the pico-second laser pulse. The photons which came through one of two cross sections of the WLS fiber in the strip were read out by a MPPC and the photons which came through the cross section on the other side, were read out by a PMT. Figure 7 shows the MPPC response as a function of absolute amount of photons measured by photomultiplier. It indicates that the saturation of response of MPPC appears in a region of large light input. This saturation behavior is represented by; ( ( ϵn )) in Nfired = Npix 1 exp, Npix 18 19 13 131 13 133 134 135 136 137 138 139 (1) where Nfired is the number of photons detected by MPPC, Npix is the number of pixels on the MPPC, ϵ is the photon detection efficiency and Nin is the number of photons incident in the MPPC sensitive area. By fitting Equation 1 to the MPPC vs PMT curve in Figure 7, the fitting parameter Npix is determined to be 44 ± 3. The actual number of pixels is 16. We think this result shows that the photon generation in the scintillator has a time duration which allows some pixels to be active again in the allocated time window, hence allowing to be hit more than once in an event. In order to apply the correction for each channel, we use the inverse function of Equation 1. The input of the function, Nfired should be an absolute number of photons as the output signal of the relevant channel. Therefore, the ADC counts of the output signals of the strip should be converted to the number of photons. To convert the ADC counts to the number of photons, the ratio of the ADC counts par photon for each channel has to be known. An LED calibration system to get such ADC-photon ratio was embedded in the prototype module and the 5

Figure 4: Four typical MIP energy distributions on respective strips. Purple line shows fitting result by a Landau function convoluted with a Gaussian function. Blue line shows the mean of pedestal peak. 14 141 14 143 144 145 146 LED calibration runs were implemented. Although the LED calibration runs were implemented for every data acquisition days, the respective ADC-photon ratios for channels are obtained, but not for each day, as an average of a half of whole data taking in this analysis. The details for the LED calibration will be discussed in the consecutive notes which will follow this result in near future. 4.3 Energy spectra Using the set of calibration constants obtained in sub-section 4.1, the measured deposit-energy is obtained in units of number of MIPs by summing the calibrated response over all strips; E total = 3 7 l=1 s=1 E ls /c ls, () 147 where i and s are the layer and strip indices, E ls is the measured energy corrected for MPPC saturation and c ls is the calibration constant on a corresponding strip. The corrected measured 148 149 15 151 15 153 154 155 energy, E ls is obtained as the following; ( E ls = n pixr ADC photon,ls log 1 N ) ADC,ls/R ADC photon,ls, (3) n pix where n pix is the effective number of pixels and R ADC photon is the ratio of ADC counts to the number of detected photons for each strip, both n pix and R ADC photon were obtained in the sub- Section 4., and N ADC,ls is the number of ADC counts as the response of each strip. The spectra of measured deposit-energy at respective beam momenta are shown in Figure 8. These spectra include some contaminations of pions and muons. In the next sub-section, the purification of the electron events from these sample is described. 6

Figure 5: The distribution of MIP constant of 16 channels. 156 157 158 159 16 161 16 163 164 165 166 167 168 169 17 171 17 173 174 175 176 177 178 4.4 Purification of electron events The selection criteria to purify the electron events are as follows: 1. the shower maximum in the ScECAL should be upstream with respect to the th layer,. the deposit-energy on the shower maximum layer in the ScECAL should be greater than; MIPs for 1 GeV/c, MIPs for 3 GeV/c, 4 MIPs for 6 GeV/c, 8 MIPs for 1 GeV/c, MIPs for 16 GeV/c, 15 MIPs for 5 GeV/c, and MIPs for 3 GeV/c, 3. the deposit-energy on the shower maximum layer in AHCAL should be less than MIPs, 4. the deposit-energy on the most downstream layer of AHCAL should be less than.4 MIPs of AHCAL, and 5. and 6. the gravitational center of electromagnetic shower in ScECAL should be in ± 4 from center on the x, y axis. The selection criteria 1-3th are in order to reject pions, 4th is to reject muons and 5th and 6th are to reject the events which have shower leakage to the side surface of the detector. Figure 9 left shows the effect of cuts and right shows the shape of spectrum after all cuts for one of the 5 GeV/c runs. The spectrum is fitted well with a Gaussian function in the range including 9% of the function area, i.e. 1.65 σ. On the other hand, the similar distribution for the one GeV/c run shows that pions remain in lower energy region of the electron peak. We discuss this issue more in 5. and 6.. 7

PC Pulse Generator Programmable Delay Gate Generator ADC ns Gate Ch.1 Ch. Voltage Power Supply Vover : 3 V H.V. Power Supply 8 V Polarized Filter Controller MPPC PMT Laser Controller Pico-sec Light Pulser ( 4nm,Violet) Thermostatic Chamber (5 ) Figure 6: Setup of the response curve measurement using pico-second laser system, the scintillator-strip and the MPPC. 179 18 181 18 183 184 185 186 187 188 189 19 191 19 193 194 195 196 197 198 For every selection cut, we investigate the variations of the measured deposit-energy in the ScECAL prototype and its resolution when changing the cut values. The measured deposit-energy is determined as the mean value of the fitted Gaussian function. Figure shows a cut variation on the shower maximum depth in the ScECAL prototype. The variation of measured depositenergy and its resolution between the case with cut on th layer and the case without cut are less than.5% and.5%, respectively. This small variation indicates that the cut does not induce the bias from rejection the large electro-magnetic showers. For every criteria and for every beam momenta, the cut variations of measured deposit-energy and its resolution are less than.1% and 1%, respectively. 4.5 Variation of measured deposit-energy depending on Run Figure 11 shows the measured deposit-energies and its resolutions for 5 GeV/c beam runs. The error bar shows only the statistic error for each run. The fluctuation of resolutions is consistent with each error represented by the error bar. However, the fluctuation of the measured deposit-energy is too large compared with individual errors. This large fluctuation indicates an existence of a large systematic uncertainty of measured deposited energy between runs which must be investigated. To avoid this systematic uncertainty in the energy resolution, we calculate the resolution for each beam momentum as the weighted average of resolutions for respective runs. Although this method does not remove the systematic uncertainty in the measured deposit-energy, we use the same way for the calculation of the measured deposit-energy in this note. For all beam momenta, such behaviors exist. The details will be discussed in Sections 5 and 6. 8

Figure 7: The response of MPPC as the function of PMT in ADC counts. Red line shows the fitting result by Eq. 1 199 1 3 4 5 6 7 8 9 11 1 13 14 15 16 17 5 Results 5.1 Linearity of energy response Figure 1 shows the measured deposit-energy in the ScECAL prototype as the function of beam momentum. The lines are results of linear fitting. The red line and dots are for the uniform region beam incidence, the blue line and dots are for the center region beam incidence, and the black line and dots show the combined results. There is no significant difference between the results for the uniform region and the center region. Although Figure 1 shows the almost good linearity, the deviations from the linear behavior in Figure 13 shows that the linearity is not enough ( < 6% ). A possible reason of this variation is gain variation of photo-sensor by temperature change. We will apply the temperature correction in the next note. The other origins of the deviation are discussed in Section 6. 5. Resolution of energy response Figure 14 shows the resolution of the measured deposit-energy in the ScECAL prototype as the function of beam momentum. Again, the red line and dots are for the uniform region beam incidence, the blue line and dots are for the center region beam incidence, and the black line and dots show the combined results. The lines are fitting results with the function of; σ E = σ 1 constant σ stochastic Ebeam (GeV ), (4) where, E is the measured deposit-energy, σ is the standard deviation of the fitted Gaussian function, and σ constant and σ stochastic are the fitting parameter for this function, and E beam is the beam energy. Table shows obtained fitting parameters for the uniform region, center region and combined. 9

4 5 1 c c 1 5 c c c c / c 5 Counts / 7.5 MIPs e su m 1 GeV 3 GeV/ 6 GeV/ 1 GeV/ 16 GeV/ 5 GeV/ 3 GeV/ Deposit-enegy in ScECAL (MIPs) Figure 8: The energy spectra on the ScECAL after the calibration and MPPC saturation correction without any cuts to select electron events. Table : The constant numbers of the terms of resolutions for the center region, uniform region, and combined results. Region Constant(%) Stochastic(%) Center 1.59±.3 14.8±.4 Uniform 1.1±.4 15.67±.5 Combined 1.44±. 15.15±.3 18 19 1 3 4 5 6 7 8 9 3 31 The errors in Table are only statistic errors. Although the constant term for the uniform region is smaller than for the center region, the small constant term for the uniform region is rather made by the larger resolution values at small beam momenta less than 6 GeV, because of the correlation between the constant term and stochastic term of the resolution. Additionally, between the uniform region and the center region, there is almost no difference of the resolution at beam momenta greater than 1 GeV. Therefore, the performance of the center region is not worse or rather better than the performance of the uniform region. 6 Discussions 6.1 Linearity of energy response The fluctuation of the measured deposit-energy between runs induces a large systematic uncertainty in the measured deposit-energy. One of the possible reason of this fluctuation is the effect of the temperature variation of the detector. The temperature monitoring was implemented during the data taking, and overall effect of temperature variation on each channel was investigated in May 9 beam test. It changes MIP calibration constants by 3.5% per one Kelvin. However the

( M ( M # o f e v e n t s 1 1 1 3 4 # o f e v e n t s 5 1 1 5 5 s ite D e p o n er g y I P ) i n E C A L D e p o s ite n er g y i n E C A L 4 6 8 9. 1 1e+4 R MS 9 35. 8 M ean 3 73 L 3. 1 E ntri e s 9 9 f ulra t f ulra n g e N oc u t In t g e ra l n g e N ocu I P ) 3 3 5 4. 8 1 8. 1 o M ean C onsta n t 799 o 1 5. 9. 5 6 / 7 6. 5 51e + 4 R MS 14. M ean E ntri e s 6 797 a l c uthf5 a l c u t hf5 y / ndf S g i m a In te g r al 3 6 7 4 o. 5 557 3 6 78 o. 7 Figure 9: Effect of cuts on the 5 GeV events in a run (Run63533), left. Black histogram shows the energy distribution before any cuts, sky blue and purple histograms show the effect of the first cut and the 1st + nd cuts, respectively, and yellow shows the effect of 1st - 3rd cuts, but events remain with additional 4th - 6th(red) cuts are almost same as the after 1st - 3rd cuts. Right plot shows the spectrum after all cuts and the red curve shows the fitting results. 3 33 34 35 36 37 38 39 4 41 4 temperature data in September 8 had problems that it sometimes did not work or had large noise. Therefore, we are still trying to recover the temperature data. The MPPC saturation correction is applied using a Equation 3 with only one parameter; the number of the effective pixels, n pix. This number is larger than the actual number of pixels, 16, because the timing structure of incoming photons allows some pixels to become active again during the event. Since there is a possibility that the timing structure is depending on the depositenergy and the incident particle species, the behavior of MPPC saturation should be studied in more detail and the systematic uncertainty from MPPC correction should be estimated. Another possible origin of the deviation is some energy leakage toward lateral sides and the downstream. We will investigate the detail of the leakage in the near future. 6. Resolution of energy response 5 43 There is no significant difference between the energy resolution for the uniform region and the 44 center region. This result indicates that the uniformity of response of the strip scintillators has 45 been improved from the prototype used at DESY. The constant term is reduced by a factor two 46 with respect to the previous prototype. 47 The reduced χ of Gaussian fitting to the every spectra are almost around one and a few 48 runs have 1.5 but less than two. However, for one GeV/c beams, the average of the reduced χ 49 is.7 because of their leading peaks. The beam at FNAL beam test includes both pions and electrons. Although we used the Ĉerenkov counter to select particles, the electron beam includes 51 an amount of pions. Especially for 1 GeV beam, pion rejection is difficult and the samples after all 5 cuts still include pions. We confirmed that the resolution decreases as the region of Gaussian fitting 53 decreases. The systematic error from this uncertainty should be estimated and more effective cuts 54 to reject pions are needed in the next step. 11

Ratio of mean of E( varied cut/nominal cut ) 1. 1.1 1.999.998 15 5 3 Cut on Max.-energy-layer in ECAL Ratio of sigma of E( varied cut/nominal cut ) 1.1 1.5 1.95.9 15 5 3 Cut on Max.-energy-layer in ECAL Figure : Cut variation of measured deposit-energy on the shower maximum cut (left) and cut variation of the energy resolution on the same cuts (right). The nominal cut value is th layer. 55 56 57 58 7 Summary and outlook The ScECAL beam test has been performed using pion - electron mixed beam in the momentum range 1-3 GeV/c. As a result of the first-stage analysis, the energy resolution combined the uniform region and the center region is found to be; σ constant = 1.44 ±.% (5) σ stochastic = 15.15 ±.3% (6) 59 6 61 6 63 64 65 66 67 68 69 7 71 7 73 Deviations from linear behavior of the energy measurement are found to be less than 6%. However, these results are very preliminary because they are estimated without any systematic uncertainties. We already found some behaviors suggesting the presence of systematic uncertainties. Several effects need to evaluate these uncertainties and they are listed in the following; Variation by the temperature should be estimated and corrected. The pion contamination should be reduced with more effective selection criteria and the systematic uncertainty from the pion contamination should be estimated. More precise MPPC saturation correction should be studied and the systematic uncertainty from the correction is needed. Significant variation of the deposit-energy in runs should be understood and corrected. The effect of wide distribution of MIP calibration constants on the energy resolution and the systematic uncertainty from this variation should be studied. The energy leakage and the uncertainty from the leakage should be estimated. To study these problems, a comparison of the data to the results of a realistic Monte Carlo simulation is necessary. 1

g y i n 1 3 4 5 6 7 ( M e so l u t i o L I P ) n D e p o s i t e n e r E CA E n e r g y r ( % ) 1 3 4 5 6 7 3 4 3 6 3 8 3. 3. 4 3. 6 Run 53533 Run 5353 Run 53531 Run 535 Run 53519 Run 53518 Figure 11: Run variations for 5 GeV/c runs. Left shows the measured deposit-energies with their statistic uncertainties. Although each uncertainty is represented with a horizontal bar, it is too small to appear in the plot. Right shows their resolutions. 74 75 76 77 78 79 8 81 8 83 84 85 86 87 88 89 9 91 9 93 94 95 96 Moreover, we took the other data set in both September 8 and May 9 for the studies as the following; to study the influence of the incident angle of particles, tilted detector runs were implemented, to study the performance on two particle separation, neutral pion runs were implemented, to study the detail position dependence of response, the beam position was precisely changed in some runs, and to study the performance of the combined detector, ScECAL, AHCAL, and Tail catcher, charged pion beam runs were implemented. Analyses for these data set are ongoing or will be started. References [1] ILD Concept Group, The International Large Detector Letter of Intent (9). [] CALICE analysis note 5. The scintillator ECAL beam test at DESY, 7 - First results. http://www.hep.phy.cam.ac.uk/ drw1/analysisnotes/can-5/can-5.pdf [3] CALICE analysis note 6. The scintillator ECAL beam test at DESY, October 9, 7 - Update 1. http://www.hep.phy.cam.ac.uk/ drw1/analysisnotes/can-6/can-6.pdf [4] CALICE analysis note 7. The scintillator ECAL beam test at DESY, April 8, 8 - Update. http://www.hep.phy.cam.ac.uk/ drw1/analysisnotes/can-7/can-7.pdf [5] CALICE analysis note 1. The scintillator ECAL beam test at DESY, June 4, 8 - Update 3. http://www.hep.phy.cam.ac.uk/ drw1/analysisnotes/can-1/can-1.pdf 13

Deposit energy in ECAL (MIP) 5 4 3 Slope = 145.8 ±.1 ( combined ) Slope = 147.83 ±.1 ( center ) Slope = 14.37 ±.1 ( uniform ) 3 Beam momentum ( GeV/c ) Figure 1: The measured deposit-energy in the ScECAL as the function of the beam momentum. The lines show the results of linear fitting with forced zero offset. The red line and dots are for the uniform region beam incidence, the blue line and dots are for the center region beam incidence, and the black line and dots show the combined results. 97 [6] Private discussion with the run coordinator of the beam control room of FNAL. 14

Deviation from linear (%) 15 5-5 - -15-3 Beam momentum ( GeV/c ) Figure 13: momentum. Deviation from the linear behavior of the measured deposit-energy as the function of the beam / E (%) σ E 18 16 14 1 8 6 4 const. = 1.44 ±. ( combined ) stat. = 15.15 ±.3 const. = 1.59 ±.3 ( center ) stat. = 14.8 ±.4 const. = 1.1 ±.4 ( uniform ) stat. = 15.67 ±.5.5 1 1/ P beam ( GeV -1/ /c -1/ ) Figure 14: The resolutions of ScECAL response of energy as the function of beam momentum. The red curve and dots are for the uniform region beam incidence, the blue curve and dots are for the center region beam incidence, and the black curve and dots show the combined results. 15